PREDICTION OF ROBOTIC MOBILE PLATFORM MOTION IN TWO-DIMENSIONAL SPACE WITH CONSIDERATION OF DYNAMIC OBSTACLES

Abstract

This work is devoted to the development of a method for predicting the route of a mobile robot and modeling the environment using the Python language. Objective. Development of a method for predicting mobile robot trajectories in an environment with dynamic obstacles, ensuring adaptive real-time routing and collision avoidance. Methodology. The research is based on mathematical modeling of mobile platform movement in two-dimensional space using a modified D* Lite algorithm. The modeling approach represents the environment as a discrete grid that includes both static and dynamic obstacles. Static obstacles simulate fixed objects such as walls or other immobile elements, while dynamic obstacles are defined by their changing position, speed, and direction of movement. Vector representation of motion is used to predict object paths based on their current state. In particular, the analysis focuses on parameters such as position, velocity, and direction of movement. The proposed statistical criterion involves selecting a stable route by examining the frequency of vertices within the range of potential trajectories. Science novelty. The method of mobile robot path planning in a dynamic environment has been improved through careful evaluation of both static and dynamic obstacles. The mathematical structure used for predicting object movement has been expanded using the incremental D* Lite method. Additionally, a new statistical criterion for optimal trajectory selection has been introduced, which takes into account path stability. Conclusions. The developed methodology enables navigation of a mobile platform in two-dimensional space with moving obstacles. It offers adaptive trajectory planning, responding to changes in obstacle positions to avoid collisions. Software modeling confirmed the algorithm's ability to compute routes in real-time. Different grid sizes were tested: 20 × 20, 50 × 50, 75 × 75, 100 × 100. Further research should be conducted in the direction of justifying the optimal size of environment mapping grids for the mobile platform, grid cell size in terms of required reaction time and positioning accuracy by the navigation system.